Spiritual But Not Religious? Evidence for Two Independent...
Transcript of Spiritual But Not Religious? Evidence for Two Independent...
Spiritual But Not Religious? Evidence for Two
Independent Dispositions
Gerard Saucier1 and Katarzyna Skrzypinska2
1University of Oregon2University of Gdansk, Poland
ABSTRACT Some psychologists treat religious/spiritual beliefs as aunitary aspect of individual differences. But a distinction between mysti-cism and orthodox religion has been recognized by scholars as well aslaypersons, and empirical studies of ‘‘ism’’ variables and of ‘‘spirituality’’measures have yielded factors reflecting this distinction. Using a largesample of American adults, analyses demonstrate that subjective spiri-tuality and tradition-oriented religiousness are empirically highly inde-pendent and have distinctly different correlates in the personality domain,suggesting that individuals with different dispositions tend towarddifferent styles of religious/spiritual beliefs. These dimensions have lowcorrelations with the lexical Big Five but high correlations with scales(e.g., Absorption, Traditionalism) on some omnibus personality inven-tories, indicating their relevance for studies of personality.
Beliefs about religious or spiritual phenomena have importanteffects on human behavior and functioning. They can provide one
with a cognitive map of the world that makes it meaningful. Suchworldview beliefs can fill many functions. They provide a paradigm
for, among other things, how the universe began, what the purposeof life is, and how to understand injustice and death (Argyle & Beit
Hallahmi, 1975); they may provide a buffer against mortality-basedanxiety, enhancing a sense of safety and security (Greenberg,
Work on this article was supported by Grant MH-49227 from the National Institute of
Mental Health, U.S. Public Health Service. The authors are grateful to Tarik Bel-
Bahar and Lewis R. Goldberg for helpful suggestions.
Correspondence regarding this article may be addressed to: Gerard Saucier,
Department of Psychology, 1227 University of Oregon, Eugene, OR 97403. E-mail:
Journal of Personality 74:5, October 2006r 2006, Copyright the AuthorsJournal compilation r 2006, Blackwell Publishing, Inc.DOI: 10.1111/j.1467-6494.2006.00409.x
Pyszczynski, & Solomon, 1986), and they may satisfy needs for
a purpose in life, anchoring a sense of what is right andwrong (Baumeister, 1991). Moreover, such beliefs connect people,
enabling the sharing of a system of values and rules that is obligatoryfor a social group (Kuczkowski, 1993), values and rules that may be
a prime guiding force for actual behavior (Ma)drzycki, 1996).Perhaps because of the way it performs these functions, religious-
ness appears to have some positive effects on health and longevity(Kozielecki, 1991; Powell, Shahabi, & Thoresen, 2003). Theseinclude protective effects with respect to alcohol/drug abuse (Miller,
1998). Nonetheless, there may be negative effects as well as positiveones (Koenig, 1997).
Despite their impact, religious or spiritual beliefs have long been amatter of only peripheral concern to personality psychologists
(Emmons, 1999). In part, this stems from the assumption thatdifferences in religiosity are a result of environmental rather than
genetic causes and might, therefore, be the proper domain ofsociologists. However, recent studies in behavior genetics suggest
that religiosity (though not denominational affiliation) is substan-tially heritable by mechanisms independent of commonly studiedpersonality traits (D’Onofrio, Eaves, Murrelle, Maes, & Spilka,
1999; Kendler, Gardner, & Prescott, 1997; Waller, Kojetin, Bou-chard, Lykken, & Tellegen, 1990), as may be true of attitudes more
generally (Olson, Vernon, Harris, & Jang, 2001). Religious experi-ence may be associated with specific aspects of brain function
(Newberg & d’Aquili, 2000; Newberg et al., 2001), and beliefs mayplay a physiological role in affect regulation (McGuire, Troisi,
Raleigh, & Masters, 1998). Such findings erode the division betweenreligiosity and other personality differences.
Psychologists outside the specialized discipline of psychology of
religion often treat religious/spiritual beliefs as a unitary aspectof individual differences. But laypersons seem able to recognize
distinct vectors in such beliefs (Zinnbauer et al., 1997). For example,one increasingly encounters phrases like ‘‘spiritual but not reli-
gious.’’ This phrase forms the title of a recent scholarly book (Fuller,2001) that discusses contemporary metaphysical religion and ‘‘un-
churched,’’ eclectic, and ‘‘psychological’’ spirituality. To a tradition-oriented adherent of a religion, such forms of spirituality may look
like one is making up one’s own faith or creating a personallycustomized worldview. Nonetheless, Fuller estimates that 20% of
1258 Saucier & Skrzypinska
Americans adhere to such unchurched spirituality, which has a long
legacy in American history.
Defining Key Terms
One can find many definitions of religiousness in the psychologicalliterature. There are concrete, abstract, metaphysical, prescriptive,
relationship-oriented, inner-motivation-oriented, and existential-quest-oriented definitions (Zinnbauer, Pargament, & Scott, 1999).
Argyle and Beit-Hallahmi (1975) defined religion as ‘‘a system ofbeliefs in a divine or superhuman power, and practices of worship or
other rituals directed towards such a power’’ (p. 1). The emphasis onworship and rituals implies community activity that binds or tiespeople together. Indeed the word religion comes from Latin religio,
derived from ligomeaning ‘‘to tie or bind’’ (etymologically related tothe English word ‘‘ligament’’). Definitions of spirituality usually put
more emphasis on the individual and on subjective experience. Theword comes from Latin spiritus, in turn from spirare (to breathe;
Wulff, 1997). Shafranske and Gorsuch (1984) defined spirituality,broadly, as ‘‘a transcendent dimension within human experience . . .
discovered in moments in which the individual questions the mean-ing of personal existence and attempts to place the self within abroader ontological context’’ (p. 231). Vaughan (1991) provided a
useful, more specific, definition: ‘‘a subjective experience of thesacred’’ (p. 105). In line with this more specific usage, in this article
we will use the more precise term subjective spirituality. In America,virtually all religious people call themselves spiritual, as do some
nonreligious people. Subjective spirituality should be understood asa narrower and less inclusive and ambiguous notion than spiritual-
ity, one closer in meaning to the natural-language term mysticism.Reports of mystical experiences are not found only among the
conventionally religious. In the general population, the tendency tomake such reports is correlated with the tendency to report para-normal experiences (Thalbourne & Delin, 1994). A variety of studies
indicate that mystical experiences are more frequently reported byindividuals who self-identify as spiritual rather than religious
(Spilka, Hood, Hunsberger, & Gorsuch, 2003). We suggest, how-ever, that the termmystical is more distinct in meaning from religious
than is the term spiritual. ‘‘Spiritual but not religious’’ likelyindicates mystical preferences, but because spiritual has more
Spiritual But Not Religious 1259
favorable connotations in English than does mystical, spiritual is
more attractive for self-description.
Tradition-Oriented Religiousness and Subjective
Spirituality in Previous Literature
Reflection on these key terms influenced us to distinguish between
tradition-oriented religiousness (TR) and subjective spirituality (SS).We find this distinction implicit in past scholarship and research.
Zinnbauer, Pargament, and Scott (1999) made specific contrastsbetween (a) organized religion and personal spirituality, (b) sub-stantive religion and functional spirituality, and (c) negative reli-
giousness and positive spirituality. In each of these contrasts, the firstterm reflects TR and the second term SS.
Emmons (1999) and others (e.g., Skrzypinska, 2002, in press) havemade the contrast in another way, noting that spirituality can be
strongly related to religiousness, though it is not always. ForEmmons, spirituality involves ‘‘a search for meaning, unity, con-
nectedness to nature, humanity, and the transcendent’’ (Emmons,1999, p. 877), thus having a strong subjective element. Religion, in
contrast, provides a ‘‘faith community with teachings and narrativesthat enhance the search for the sacred and encourage morality’’(Emmons, 1999, citing Dollahite, 1998, p. 877), thus having a strong
traditionalist element. Emmons’s contrast is, we believe, moreprecisely expressed as one between subjective spirituality and tradi-
tion-oriented religiousness.Empirical support for this contrast is provided by a recent study
of the content of isms terms in the English language. Saucier (2000),assuming that the most important worldview-belief concepts tend to
become represented in words ending in ism, extracted 266 such termsfrom an English-language dictionary and built items directly fromtheir definitions. In a sample of 500 college students, the replicable
item structures had no more than four factors. A parallel study ofisms found in a Romanian-language dictionary, in Romania,
replicated the four-factor structure (Krauss, in press). Two ofthe four factors—Alpha and Delta—are relevant here. Concepts
loading most highly on Alpha emphasize individual differences inadherence to traditional and religious sources of authority; this
factor is correlated substantially with conservatism and authoritar-ianism. Concepts loading highly on Delta emphasize unorthodox
1260 Saucier & Skrzypinska
spirituality. They involve individual differences in beliefs emphasiz-
ing intuition and spiritual experiences of a mystical nature, butincluding some currently fashionable superstitions. Alpha and Delta
appear to encapsulate the distinction between tradition-orientedreligiousness and subjective spirituality.
A set of factors corresponding to Alpha and Delta was reportedby MacDonald (2000), who sought to identify the common dimen-
sions in 11 prominent measures of religious and spiritual constructs.He found five factors: Religiousness, Cognitive Orientation Towards
Spirituality (COTS), Experiential/Phenomenological (E/P), Paranor-mal Beliefs, and Existential Well-Being. A higher-order factoranalysis found two factors. One was labeled Cognitive and Beha-
vioral Orientation Towards Spirituality and included Religiousnessand COTS. The other factor was labeled Non-Ordinary Experiences
and Beliefs; capturing distinctions in previous measures of mysti-cism, it included E/P and Paranormal Beliefs. Existential Well-Being,
whose content overlaps with Emotional Stability versus Neuroti-cism, did not have appreciable loadings on either higher-order factor.
These two higher-order factors appear to correspond to TR and SS.Outside psychology, scholars describing diverse religions have
noted a distinction between mystical schools of thought on the one
hand and more orthodox trends on the other (e.g., Sabatier, 1905;Schuon, 1953), fundamentalism being in many respects an attempt to
reassert orthodoxy. For example, in Islam, Sufism represents amystical school, whereas Islamic orthodoxy is better represented in
the Salifi or Wahhabi schools of practice. In Judaism, the mysticalKaballa contrasts with various orthodox schools. Zen is an espe-
cially mystical form of Buddhism.The terms esoteric and exoteric are sometimes used to capture
these distinctions (Schuon, 1953). According to Schuon, exotericreligion emphasizes form and tends toward literalistic dogmas, aclaim to exclusive possession of the truth, sentimentality, and an
emphasis on morality and personal salvation (which serves indivi-dual interest, such as reward or relief in an afterlife), couched in ways
that make it attractive to a majority of people. Esoteric religion, incontrast, is more metaphysical, contemplative, oriented to knowl-
edge, wisdom, and unification with divinity, and toward the spiritand not the letter of religious teachings. A similar distinction was
made by Sabatier (1905) between authority- and spirit-focusedapproaches to religion.
Spiritual But Not Religious 1261
The terms esoteric and exoteric should not be confused with the
terms intrinsic and extrinsic used by Allport (1959). These do notconcern belief but rather motivation: Intrinsics have high commit-
ment to religious activities and beliefs, treating religion as an end initself; extrinsics use religion as a means to desired personal ends (e.g.,
status, comfort). Intrinsic and extrinsic distinguish two ways of beingreligious and are concepts that were envisaged to describe individual
differences within religious populations; indeed, the distinction hasbeen problematic to apply outside such populations (e.g., Burris,1994). TR and SS, in contrast, are dimensions applicable to general
population samples, although one might conceivably identify bothintrinsics and extrinsics among individuals high in either dimension.
Hypotheses
We made the empirical conjecture that the distinction between
mystical and orthodox forms of belief corresponds to that oftenmade between the terms spiritual and religious and with individual
differences in important psychological variables. We set out toinvestigate two questions: Is the division of religious/spiritual beliefs
into two independent dimensions a psychometric reality? And arethe correlates with psychological variables so distinct and substantialthat psychologists should be paying attention to these dimensions?
Our hypotheses all follow from an overarching conception of thedifference between TR and SS. TR involves a reliance on trusted
sources of authority (such as scriptures or a church) that are a sharedreference point for a group of individuals—these sources providing
clarity but also an impetus for conformity. SS is a more subjectivistand individualized approach that involves more reliance on private
imagination and intuition, more egalitarianism, more nonconfor-mism, and more of a questioning attitude toward status-quocollective norms.
A first hypothesis was that in English the term spiritual differsfrom the term religious in a way that corresponds to the contrast
between SS and TR. This is in line with empirical findings ofZinnbauer et al. (1997), that is, religiousness being relatively
more highly associated with authoritarianism, orthodoxy, andchurch attendance; spirituality being relatively more associated
with mystical experiences and New Age beliefs and practices. Wealso hypothesized that the term mystical would correspond more
1262 Saucier & Skrzypinska
closely to the SS side of the contrast, and that responses to the single
adjectives mystical and religious would be approximately orthogonalin self-ratings and would correspond with the two higher-order
factors of spirituality found by MacDonald (2000). We furtherhypothesized that these two sets of orthogonal distinctions would
correspond to that orthogonal distinction between Alpha and Deltafactors emerging in analyses of isms terms (Krauss, in press; Saucier,
2000). If all these initial hypotheses were supported, it would bepossible to conceive of two latent, highly independent dimensions
having three sets of indicators, one set from the adjectival lexicon,one based on factors from spirituality measures, and one based onfactors from isms concepts.
In line with our overarching conception, we formed a variety ofhypotheses regarding how TR and SS would diverge with respect to
substantive associations with other variables.We hypothesized that TR (and not SS) would be related to the
following: individual differences in traditionalism, authoritarianism,collectivism, and impression management; attitudes emphasizing the
power of divinity hierarchically exercised (e.g., by a supernaturalGod through miracles); believing it important to respect whateverrepresents the sources of such divine authority (e.g., scriptures,
religious rules, and leaders) while not respecting those whose beliefsor behavior go against such authority (e.g., evolutionists, gays,
feminists); and, behaviorally, high engagement in religious practicesand low levels of drug and alcohol use.
We hypothesized that SS (and not TR) would be related to thefollowing: individual differences in absorption, fantasy-proneness,
dissociation, private self-consciousness, eccentricity, (low) social-dominance orientation, and individualism (especially of an egalitar-
ian variety); attitudes emphasizing the power of nonhierarchicalsupernatural forces (e.g., magic, witchcraft, astrology, fate); believ-ing it is important to respect those aligned with such forces (e.g.,
enlightened persons and psychics); and, behaviorally, high engage-ment in environmentalist practices and the pursuit of personal
creative achievement.We hypothesized that ‘‘personality’’ scales designed to capture
both religious and spiritual tendencies (e.g., self-transcendence)would correlate with both TR and SS. Correlated in opposite
directions with TR and SS would be some other variables, includingthose related to conformity (vs. nonconformity), favoring (vs.
Spiritual But Not Religious 1263
questioning) of norms, hierarchical relations of authority, and the
contrast between collectivism and individualism.Religiousness appears to be generally independent of the Big Five
(Piedmont, 1999; Saucier & Goldberg, 1998). However, a trend inprevious studies (reviewed by Saroglou, 2002) indicates that religi-
osity is correlated, although quite modestly, with the Big Five factorsof Agreeableness and Conscientiousness. We suggest that such
relations depend on the content of the religiousness measure, inparticular how much the measure emphasizes TR as compared to SS.We hypothesized that TR would be related to Agreeableness and
Conscientiousness but that SS would be associated instead withOpenness to Experience (and with its lexical-factor counterpart,
Imagination/Intellect).
METHOD
Participants
The sample consisted of those 375 members of the Eugene-Springfieldcommunity sample who completed all measures described below andincluded 160 men and 215 women, with an average age of 51.23 (standarddeviation 12.4) at first recruitment in 1993 (by way of response to amailing sent to area homeowners). Most measures administered to thissample have involved personality and not beliefs. Those used in this studywere administered between 1993 and 2002.
Responses to a multiple-choice question about denominational affilia-tion were available for 358 of the respondents (the other 17 respondentshad not returned an entire survey questionnaire containing this item). Ofthe 358, 44% indicated a Protestant affiliation, 9% Catholic, and 2%each for Mormon and Jewish. Corresponding exactly to Fuller’s (2001)estimate for ‘‘unchurched spirituality’’ in America, 20% endorsed ‘‘spiri-tual, but not affiliated with a conventional religion.’’ Another 6%endorsed ‘‘other,’’ and 16% endorsed a ‘‘none’’ option. Participantswere included in analyses regardless of their category and degree ofreligious affiliation.
The cross-time stability of religious attitudes in older adults is quitehigh. For example, in this sample the retest correlation between self-description responses to the single adjective Religious in 1993 and 1996was an impressive .80—higher than is typically expected for personalityitems and scales. And, as will be seen, response to this single item in 1993correlated nearly .80 with a nonadjectival index of religiosity adminis-tered 9 years later.
1264 Saucier & Skrzypinska
Measures
Adjectives. The term Religious (from the 1993 administration, whichhad a higher N than that of 1996) was used (1–9 rating scale), as were theterms Mystical and Spiritual (administered in 2002, with a 1–7 ratingscale). The term Religious was not readministered in 2002.
ESI items. Items from MacDonald’s (2002) Expressions of SpiritualityInventory (ESI) were used, except for the omission of items from thismeasure’s well-being scale, a fairly conventional measure of Neuroticism(of which we had other measures), whose items lack explicit spiritual orreligious content. The remaining 24 items, all of which included explicitmention of spiritual or religious content, were factor analyzed by twomethods (principal components and maximum likelihood extraction, ineither case with both oblimin and varimax rotation), extracting twofactors, which were extremely similar across methods. Because of ourinterest in retaining exact factor scores, we relied henceforth on theoblimin-rotated components, which correlated .36. As expected, thesetwo dimensions corresponded well to TR and SS. The two componentswere interpreted as Religiousness (with content referencing attendingservices and the cognitive importance and effects of religion in one’s life)and Spiritual Experiences (with content referencing experiences that aremystical, transcend space and time or the usual sense of self, and alsoparanormal beliefs regarding psychokinesis, ghosts, predicting the future,leaving one’s body, and communicating with the dead).
Alpha and Delta isms factors. Saucier (2004) described the developmentof the Survey of Dictionary-Based Isms (SDI), which consists of 48 itemclusters (of roughly four items each) centered around one or more ismsconcepts found in an English-language dictionary. When these 48 itemclusters were factor analyzed, in the same community sample used here,four obliquely rotated factors were virtually orthogonal and nearlyidentical to the four varimax factors, which corresponded well to thefour factors from earlier studies of dictionary-based isms (Saucier, 2000).We used the factor scores for Alpha and Delta (from the analysis of 48clusters) in our analyses.
Supplementary attitude scales. Saucier (2004) also described the devel-opment of 42 additional item clusters representing constructs from theprevious literature not directly referenced in the SDI or otherwisecomplementary to SDI content. In this study, we used three of thesescales: those for Quest-Orientation (a questioning way of being religious;coefficient Alpha [a] .80), Hierachialism (valuing obedience/deference to
Spiritual But Not Religious 1265
those in a higher social position; a .55), and Extropunitiveness (hier-archically exercised harshness toward criminals; a .69).
Eccentricity. We utilized an unpublished scale developed by Goldbergwith 21 items (a .84) from the International Personality Item Pool (http://www.ipip.ori.org). Example items are ‘‘Know that my ideas sometimessurprise people,’’ ‘‘Am able to disregard rules,’’ ‘‘Love to dress inoutlandish clothes,’’ and (reverse-scored) ‘‘Like to be viewed as properand conventional.’’
Conformity. The adjectives Nonconforming and Conforming were admi-nistered to participants, and a Conforming-minus-Nonconforming scorecalculated (a .92). We expected TR to be positively and SS to benegatively related to this index. Norm-Favoring vs. Norm-Questioning(v. 2; a .74) from the California Psychological Inventory (CPI; Gough,1996) was used as an additional index of conformity.
Impression management. Another possible aspect of conformity is theneed for approval (Crowne & Marlowe, 1964), now best represented inmeasures of impression management (IM; Paulhus, 1984). We used theBalanced Inventory of Desirable Responding (Paulhus, 1988) IM scale,with continuous, not dichotomized, scores (a .82). A previous studyindicated a significant correlation between religiosity and impressionmanagement (Gillings & Joseph, 1996).
Private self-consciousness. Private self-consciousness is a disposition tobe highly aware of internal states (Fenigstein, Scheier, & Buss, 1975). Weadministered the scale of Buss (1980; a .76).
Authoritarianism. We used a selection of Right-Wing Authoritarianism(RWA) items (Altemeyer, 1996). The 14 items (seven pro-trait, seven con-trait; a .91) were selected as a group that maximized coverage of thecontent in the longer scale (as found in stepwise regression of RWA itemson RWA score in the Study 2 data set from Saucier [2000]), whilemaintaining balanced keying of pro-trait and con-trait items.
Social dominance orientation (SDO). We used a selection of SDO items(from Pratto, Sidanius, Stallworth, & Malle, 1994). The eight items (fourpro-trait, four con-trait; a .80) were selected as a group that maximizedcoverage of the content in the longer scale (as found in stepwise regressionof SDO items on SDO score in the Study 2 data set from Saucier [2000]),while maintaining balanced keying of pro-trait and con-trait items. We
1266 Saucier & Skrzypinska
hypothesized that SS, due to its egalitarian or ‘‘horizontal’’ emphases,would be negatively associated with Social Dominance Orientation.
Collectivism and individualism. We used scales by Triandis and Gelfand(1998). These have ‘‘horizontal’’ (egalitarian) and ‘‘vertical’’ (pro-hier-archy) subscales for both Individualism (idiocentrism) and Collectivism(allocentrism). In keeping with our hypotheses, we aggregated the twoCollectivism subscales (a .68) and for Individualism used only thehorizontal Individualism subscale (a .57). We also used the Oyserman(1993) scales for Individualism (a .55) and Collectivism (a .67) andsubtracted scores on one from the scores on the other to arrive at aCollectivism-minus-Individualism score (a .63).
From the Magical Ideation Scale of Eckblad and Chapman (1983) weused all 30 items but changed the response format from true-false to a 5-point format (strongly agree, agree, not sure, disagree, strongly disagree).The scale had a coefficient Alpha of .92.
Irrational beliefs (Superstitiousness). We used the 19-item scale ofKoopmans, Sanderman, Timmerman, and Emmelkamp (1994) withcoefficient Alpha of .93. Content includes beliefs in psychokinesis, out-of-body experiences, astrology, reincarnation, spells, and psychic powers.
Fantasy-proneness. We used the Creative Experiences Questionnaire(CEQ; Merkelbach, Horselenberg, & Muris, 2001). However, we sepa-rated out the eight ‘‘as a child’’ items as a measure of childhood fantasies(via retrospective recall; a .72) because they give a retrospective accountof childhood fantasy-proneness, a potentially useful antecedent variable.The 17 remaining items were taken as a measure of fantasy-proneness inadulthood (a .77). The two measures (current and retrospective) corre-lated .41.
Dissociation. We utilized the 31-item Curious Experiences Survey (CES;Goldberg, 1999), a revision of the Dissociative Experiences Scale (Bern-stein & Putnam, 1986). Alpha was .90.
MPQ scales. We used scores from the Traditionalism (a .87) andAbsorption (a .90) scales from the Multidimensional Personality Ques-tionnaire (MPQ; Tellegen, in press).
The Temperament and Character Inventory (TCI; Cloninger et al.,1994) has a Self-Transcendence scale (a .95) with five subscales, all ofwhich were used in our analyses. The labels for the subscales in order(ST1 to ST5; a .79, .77, .90, .95, .82) are: Self-forgetful versusSelf-conscious experience (i.e., absorption), Transpersonal identification
Spiritual But Not Religious 1267
versus Self-differentiation (e.g., sense of connection with all things),Spiritual acceptance versus Rational materialism (e.g., contact with anddirection by a higher power), Enlightened versus Objective (supernaturalguidance), and Idealistic versus Practical (e.g., engagement with prayerand moral ideals). We hypothesized that some of these subscales would becorrelated mainly with TR, the others with SS, and, by implication, thatST subscales can be differentiated based on relative TR and SS loadings.
Big Five. To index the Big Five factor structure we used the Mini-Markers (Saucier, 1994) to capture the lexical representation of thestructure and the NEO Personality Inventory-Revised (Costa & McCrae,1992) to capture its best-known questionnaire representation. Coefficientalpha values ranged from .77 to .86 for Mini-Markers and .85 to .92 forthe NEO-PI-R.
Multi-Language Seven. To index a variant factor structure that is alsolexically derived but may provide a better fit to indigenous factorstructures from non-European languages, we used the 60ML7, a 60-adjective measure of the Multi-Language Seven factors (Gregariousness,Self-Assurance, Even Temper, Concern for Others, Conscientiousness,Originality/Intellect, Social Unacceptability/Negative Valence) withalpha values of .70 to .81. These seven factors were derived fromcommonalities found in structures from lexical studies of Filipino(Church, Reyes, Katigbak, & Grimm, 1997) and Hebrew (Almagor,Tellegen, & Waller, 1995), studies in which Big Five structures were notobtained (Saucier, 2003).
Behavioral frequency reports. We used four clusters developed by Gold-berg (in press) from a set of 400 activity descriptions; participants ratedthe relative frequency with which they engaged in the activity. Theclusters involved drug/alcohol behavior (a .89; e.g., became intoxicated,smoked marijuana, drank beer), religious practices (a .87; e.g., read theBible, taught Sunday school, prayed), creative achievement behavior (a.70; e.g., produced a work of art, wrote poetry, acted in a play), andproenvironmentalist practices (a .76; e.g., rode a bicycle to work, boughtorganic food, recycled).
Attitude about categories of people. To tap specific attitudes aboutclasses and groups of people, 32 items were administered with thequestion ‘‘Who deserves more influence, power, and respect?’’ and a 5-point response scale (much less, slightly less, already have the right amount,slightly more, much more). Items suiting our hypotheses were: religious
1268 Saucier & Skrzypinska
leaders, those who study the holy scriptures, those who follow the TenCommandments, military leaders, gay and lesbian people, scientists whobelieve in evolution, feminists, those with psychic abilities, those withunconventional spiritual beliefs, spiritually enlightened persons, andpeople who avoid military service.
Attitudes about causal forces. To tap other consequential specific beliefs,44 items were administered with the question ‘‘What forces determinewhat happens in life?’’ and a 5-point response scale (never, seldom,sometimes, often, always). Items suiting our hypotheses were: God,miracles, supernatural power, astrological influences, magic, witchcraft,destiny, and fate.
Childbearing and divorce. Reflecting the likely greater collectivism of TRand individualism of SS, we hypothesized that TR would be associatedwith a lower likelihood of divorce and a larger number of children,whereas SS would be associated with a higher likelihood of havingexperienced a divorce and with smaller numbers of children. Numberof biological children and frequency of divorce were elicited with twoitems from among a larger 18-item survey, headed ‘‘Things That Don’tHappen Every Day,’’ that asked respondents to identify how often eachof 18 events had occurred in their lives. The two items used here were‘‘gave birth to or fathered a child’’ and ‘‘been divorced.’’ The 6-pointresponse scale had response options of never, once, twice, three times, fourtimes, and five or more times. For this sample the median number ofchildren was two, with 14% percent having never begotten a child. Some33% of the sample had experienced at least one divorce; responses toanother item indicated that only 5% had never been married.
RESULTS
The three sets of indicators for TR and SS were not significantly
correlated with the age of the participant—correlations with ageranging from � .05 to .10 for TR indicators and from � .10 to .07for SS indicators. All of the indicators, on the other hand, had a
very modest point-biserial correlation with gender. With positivecorrelations indicating higher scores among women than men, the
correlations were .12 (Religious), .21 (ESI Religiousness), .14 (SDIAlpha), .09 (Mystical), .21 (ESI Spiritual Experiences), and .26 (SDI
Delta). We infer that TR and SS are both unrelated to age, but bothappear to be weakly related to being female rather than male.
Spiritual But Not Religious 1269
Response means (on a 1–7 scale) were lower for Mystical (2.87;
SD 1.78), than for Religious (4.28; SD 1.94) and for Spiritual (4.61;SD 2.19). This is in harmony with Norman’s (1967) report indicating
that, in English, on a 1 to 9 scale, Mystical (M5 5.22) is less sociallydesirable than Religious (M5 6.57) and Spiritual (M5 6.42).
The correlations among these terms were .68 (Religious-Spiritual),.26 (Spiritual-Mystical), and .09 (Religious-Mystical). Controlling
for Religious (in a partial correlation analysis) did not affect thecorrelation between Spiritual and Mystical. Indications are, then,that Spiritual is somewhat intermediary between the other two
concepts, whereas Religious and Mystical have more independentdenotation. Spiritual and Religious were highly correlated, but,
nonetheless, when a Spiritual-minus-Religious score was calculated,this score also correlated positively (.26) with Mystical in the present
sample, consonant with our predictions.Table 1 presents all correlations among the three sets of putative
indicators for TR and SS. We note that the within-set intercorrela-tions for TR are very high, much higher than those for SS, indicating
a more tightly converging construct. Between-set correlations (TRwith SS indicators) were generally small. Because so many hypoth-eses were tested simultaneously, we used a relatively stringent
criterion ( po.001) for statistical significance in this and other tables.By far, the largest between-set correlation (.36) was between ESI
Religiousness and ESI Spiritual Experiences. Scrutiny of the 24 ESIitems revealed that six use the term spiritual, whereas only two use
religious and one uses mystical. The items that included the termspiritual all had at least moderate (.37 or higher) loadings on both
the Religiousness and Spiritual Experiences factors. Moreover, itemsreferring to belief in the reality of witchcraft and of spirits and ghostsand to merger with a ‘‘force or power greater than myself’’ also had
substantial loadings on both factors. Avoiding these items that mixTR and SS content, we did identify small subsets of ESI items that
could be used as adequately reliable abbreviated indicator scales forTR and SS with a near-zero correlation; the six items involved
reference to seeing oneself as a religiously oriented person and to theimportance of going to religious services (for TR; a .87), as well as
transcending space and time, leaving one’s body, communicatingwith the dead, and psychokinesis (for SS; a .78). However, because
our particular selection of orthogonal item-sets might capitalize onchance features of the present sample, we relied for subsequent
1270 Saucier & Skrzypinska
Ta
ble
1C
orr
ela
tio
nM
atr
ixfo
rT
hre
eSe
tso
fIn
dic
ato
rsfo
rD
ime
nsi
on
so
fT
rad
itio
n-O
rie
nte
dR
eli
gio
usn
ess
an
dSu
bje
cti
ve
Spir
itu
ali
ty
Indicator
Tradition-O
rientedReligiousness
SubjectiveSpirituality
Religious
ESIReligiousness
SDIAlpha.()
Mystical
ESISpiritualExpcs.SDIDelta.()
Religious
ESIReligiousness
.78(.72/.84)
SDIAlpha.()
.76(.69/.83)
.81(.75/.87)
Mystical
.09(�.01/.19)
.19(.09/.29)
.11(.01/.22)
ESISpiritualExperiences
.17(.07/.27)
.36(.26/.45)
.20(.10/.30)
.51(.42/.60)
SDIDelta.()
.06(�.04/.17)
.18(.08/.28)
.01(�.09/.12)
.37(.27/.46)
.52(.44/.61)
Note.N
5375.ReligiousandMysticalreferto
single
adjectives
treatedas1-item
measures.
ESI—
ExpressionsofSpirituality
Inventory
(MacD
onald,2002).
SDI—
Survey
ofDictionary-Based
Isms(Saucier,2004).
Coefficients
inparentheses
are
95%
confidence
limits.
Correlationsover
.40are
printedin
boldface
type.
analyses on the obliquely related principal components based on all
of the 24 ESI items; use of orthogonal components from the same 24items led to very similar external correlates.
Overall, we interpreted these results as evidence in support ofusing these measures—the adjectives religious and mystical, the two
isms factors, and the two ESI factors—as converging indicators foreach of two religious/spiritual belief dimensions that tend to be
highly independent of one another and can evidently be made fullyorthogonal with ease, if desired.
Given the good convergence among the indicators within each set,
we simplified the hypothesis testing by aggregating the indicators.One means to this end was a principal components analysis. When
the six indicators were analyzed with two factors extracted androtated by the oblique oblimin method, they loaded as expected on
separate TR and SS factors, which intercorrelated only .18, with a95% confidence interval from .08 to .28. Use of maximum-likelihood
extraction led to nearly identical factors, intercorrelating .22. In thecomponent solution, salient structure-matrix loadings on the first
factor were .93 (Alpha), .93 (ESI Religiousness), and .91 (Religious),whereas salient loadings on the second factor were .85 (ESI SpiritualExperiences), .79 (Delta), and .77 (Mystical), with all other loadings
being of magnitude .31 or lower. Component scores were saved fromthis analysis and compared to composite variables constructed by
standardizing all six indicators, then averaging the three TR indica-tors, and then separately averaging the three SS indicators. The
composite variables correlated .9994 (TR) and .9990 (SS) with thecorresponding component scores and .21 with each other (95%
confidence interval from .11 to .31). Treating each set of threeindicators as alternate measures of a single construct, we couldcompute precisely the internal consistency of these composite vari-
ables (a .91 for TR, .72 for SS). Therefore, we used these bypreference in our hypothesis tests (although a comparison indicated
that the nearly identical component scores would have led to thesame conclusions).
A question of some interest is whether hypotheses would havebeen supported if we had used only one kind of indicator. To enable
examination of this subsidiary issue, we provide additional correla-tions between each indicator and the measures. But we emphasize
that in Tables 2–7, the two leftmost columns of correlations are thecrucial ones. The next two columns to the right provide estimates of
1272 Saucier & Skrzypinska
the correlations after correction for attenuation due to imperfect
reliability of the measures. These corrections should be interpretedwith some caution because such corrections can be excessive to the
degree that the two variables involved are multidimensional insimilar ways (Schmitt, 1996); the scales in the tables likely vary in
their degree of multidimensionality, so there is a hazard of over-correction in some instances.
Variables Hypothesized to Relate to Tradition-Oriented Religiousness
Tables 2 through 4 present correlations between the TR and SS
variables and hypothetically related scales. Table 2 concentrates onvariables hypothesized to be related to TR. TR was strongly
correlated with Right-Wing Authoritarianism and with Traditional-ism but also moderately correlated with Collectivism and with
Impression Management. Correlations with the SS indicators wererarely significant (po.001). These results support our hypotheses.
We also found support for hypotheses involving variables basedon act-frequency self-reports. TR was very strongly positively
correlated with engagement in religious practices and was negativelycorrelated with reported engagement in drug and alcohol-relatedbehaviors.
Variables Hypothesized to Relate to Subjective Spirituality
Table 3 includes variables hypothesized to be related to SS. Absorp-
tion, magical ideation, superstitious beliefs, and fantasy-proneness inadulthood were all substantially positively correlated with SS.
Fantasy-proneness in childhood had moderate correlations withSS, as did dissociation and eccentricity. Correlations for private
self-consciousness and for social dominance orientation were sig-nificant for the aggregate variable, but only for two of the three SSindicators, taken individually. The individualism measures turned
out to be more highly correlated with TR (negatively) than with SS.It appears, then, that TR is a bipolar dimension that is partially
aligned with collectivism versus individualism, whereas SS has littlerelation to this bipolar dimension.
Reports of creative achievement behaviors had weak correlationsin the expected positive direction with SS but not consistently across
Spiritual But Not Religious 1273
Ta
ble
2C
orr
ela
tio
ns
Wit
hSc
ale
sE
xp
ec
ted
toR
ela
teto
Tra
dit
ion
-Ori
en
ted
Re
lig
iou
sne
ss
Tradit.Religiousness
Subj.Spirituality
Measure
TR
SS
TR0
SS0
Religious
ESI-R
SDI
Alpha
Mystical
ESI-S
SDI
Delta
ReligiousPractices
.80n
.01
.90
.01
.75n
.77n
.71n
.01
.11
�.10
Right-WingAuthoritarianism
.64n�.15
.70�.18
.53n
.50n
.75n
�.04
�.10
�.20n
MPQ
Traditionalism
.55n�.17n
.62�.21
.46n
.43n
.63n
�.09
�.17
�.15
Drug/A
lcoholBehavior(–)
�.44n
.10
�.49
.12
�.41n
�.38n�.42n
.04
.05
.15
Collectivism
(Triandis&
Gelfand)
.31n
.09
.43
.14
.30n
.33n
.24n
.03
�.01
.16
Collectivism
(Oyserm
an)
.24n
.02
.31
.03
.21n
.22n
.24n
.00
.00
.04
ImpressionManagem
ent
.21n�.08
.24�.10
.21n
.21n
.16
�.03
�.08
�.08
Note.N
5375.TR0 —
Tradition-orientedReligiousnesscomposite,TR0 —
this
composite
withcorrectionforattenuation;SS—
Subjective
Spirituality
composite,SS0 —
this
composite
with
correction
for
attenuation;ESI—
Expressions
ofSpirituality
Inventory;ESI-R—
Religiousnessfactor;
ESI-S—
SpiritualExperiencesfactor,
(�)indicatesthata
negativecorrelation
washypothesized.Attenuation-
correctedcoefficients
are
printedin
italics.
npo.001.
Ta
ble
3C
orr
ela
tio
ns
Wit
hSc
ale
sE
xp
ec
ted
toR
ela
teto
Sub
jec
tiv
eSp
irit
ua
lity
Tradit.Religiousness
Subj.Spirituality
Measure
TR
SS
TR0
SS0
Religious
ESI-R
SDI
Alpha
Mystical
ESI-S
SDI
Delta
Irrational/SuperstitiousBeliefs
.08
.65n
.09
.79
.00
.13
.09
.43n
.59n
.55n
MagicalIdeation
.06
.57n
.07
.71
�.03
.10
.09
.39n
.53n
.46n
MPQ
Absorption
.05
.55n
.06
.68
.03
.12
�.01
.37n
.52n
.43n
CEQ
Fantasy
Proneness
.15
.51n
.18
.68
.11
.20n
.10
.38n
.53n
.32n
Eccentricity
�.20n
.37n�.23
.47
�.20n
�.10
�.25n
.26n
.40n
.22n
CEQ
ChildhoodFantasies
�.05
.35n�.06
.48
�.06
.01
�.09
.19n
.37n
.29n
CESDissociation
.07
.34n
.08
.42
.03
.10
.06
.20n
.34n
.26n
Private
Self-Consciousness
.00
.25n
.00
.34
�.03
.06
�.03
.23n
.28n
.10
SocialDominance
Orientation(�)
.06
�.23n
.07�.30
.00
�.03
.19n�.10
�.17
�.29n
Environmentalist
Practices
�.22n
.13
�.26
.18
�.19n
�.13
�.30n
.03
.16
.13
CreativeAchievem
entBehaviors
�.06
.13
�.07
.18
�.05
.02
�.13
.07
.21n
.04
Individualism
(Oyserm
an)
�.26n
.12
�.37
.19
�.24n
�.26n�.21n
.09
.11
.11
HorizontalIndividualism
�.20n
.10
�.28
.16
�.22n
�.16
�.18n
.06
.05
.14
Note.N
5375.TR—
Tradition-oriented
Religiousnesscomposite,TR0 —
this
composite
with
correction
forattenuation,SS—
Subjective
Spirituality
composite,SS0 —
this
composite
with
correction
for
attenuation,ESI—
Expressions
ofSpirituality
Inventory,ESI-R—
Religiousnessfactor,
ESI-S—
SpiritualExperiencesfactor,
(�)indicatesthata
negativecorrelation
washypothesized.Attenuation-
correctedcoefficients
are
printedin
italics.
npo.001.
all indicators. Environmentalist practices were more consistently
related (but negatively and weakly) to TR than to SS indicators.
Variables Hypothesized to Relate to Both TR and SS
Table 4 presents results for variables hypothesized to be related toboth TR and SS. Self-Transcendence was, as hypothesized, related
to both, albeit more strongly to TR. This overall effect, however,masks interesting divergences at the subscale level. Self-Forgetful-
ness and Transpersonal Identification were both more highly corre-lated with SS, whereas Spiritual Acceptance, Enlightened, andIdealistic were more highly correlated with TR.
The last variables in Table 4 were hypothesized to be related inopposite directions to TR and SS, and thus to differentiate these two
dimensions. The best differentiator was Quest Orientation, whichwas related positively to SS and negatively to TR. ‘‘Religion as
quest’’ is characterized as ‘‘an open-ended active approach toexistential questions that resists clear-cut, pat answers’’ (Batson &
Schoenrade, 1991, p. 416), thus emphasizing individuality overtradition. Directions of effect for the other variables were allconsistent with hypotheses but not significant for both aggregate
variables.
Correlations of TR and SS With Single Items
We had numerous hypotheses relating TR or SS to attitudesregarding causal forces and categories of persons. Table 5 presents
relevant correlations. Because the attitudes were indexed with singleitems of unascertained reliability, no corrections for attenuation are
offered.As predicted, TR was strongly associated with a belief that God is
a powerful force, with a belief in the power of supernatural forces
and miracles, and with respect for religious leaders, those who studythe scriptures, those who follow the Ten Commandments, and
(negatively) scientists who believe in evolution. There were moderatepositive correlations with respect for military leaders and, negatively.
with respect for gays and lesbians and feminists.As predicted, SS was associated with superstitious beliefs, such as
in the power of astrology or magic, and with respect for psychics andbelief in the power of destiny. Several other variables hypothesized
1276 Saucier & Skrzypinska
Ta
ble
4C
orr
ela
tio
ns
Wit
hSc
ale
sE
xp
ec
ted
toR
ela
teto
Bo
thD
ime
nsi
on
s
Tradit.Religiousness
Subj.Spirituality
Measure
TR
SS
TR0
SS0
Religious
ESI-R
SDI
Alpha
Mystical
ESI-S
SDI
Delta
Hypothesized
tobeboth
TR
andSS-related
TCISelf-Transcendence
.76n
.52n
.82
.63
.67n
.79n
.64n
.31n
.56n
.38n
TCIST1:Self-forgetful
.15
.52n
.18
.69
.14
.21n
.07
.33n
.55n
.38n
TCIST2:Transpersonalidentif.
.25n
.59n
.30
.79
.22n
.33n
.15
.36n
.53n
.53n
TCIST3:Spiritualacceptance
.81n
.43n
.89
.53
.71n
.83n
.70n
.24n
.51n
.30n
TCIST4:Enlightened
.87n
.33n
.93
.40
.76n
.87n
.78n
.20n
.40n
.20n
TCIST5:Idealistic
.75n
.35n
.87
.45
.67n
.75n
.67n
.22n
.36n
.27n
Hypothesized
opposingrelationsto
TR
andSS
QuestOrientation
�.21n
.33n�.25
.43
�.17
�.11�.31n
.23n
.23n
.33n
Collectivism
vs.Individualism
(Oys.)
.32n�.07
.42�.10
.29n
.31n
.29n�.05
�.07
�.05
Hierarchialism
.31n�.05
.44�.08
.25n
.22n
.39n
.01
�.02
�.12
Extropunitiveness
.25n�.06
.31�.08
.16
.15
.38n
.04
�.08
�.10
CPINorm
-favoring(v.2)
.20n�.12
.24�.16
.20n
.20n
.17
�.07
�.13
�.09
Conform
ingversusNonconform
ing
.16�.15
.17�.18
.17
.07
.19n�.15
�.20n�.02
Note.N
5375.TR—
Tradition-oriented
Religiousnesscomposite,TR0 —
this
composite
with
correctionforattenuation,SS—
Subjective
Spirituality
composite,SS0 —
this
composite
with
correction
for
attenuation,ESI—
Expressions
ofSpirituality
Inventory,ESI-R—
Religiousnessfactor,
ESI-S—
SpiritualExperiencesfactor,
(�)indicatesthata
negativecorrelation
washypothesized.Attenuation-
correctedcoefficients
are
printedin
italics.
npo.001.
Ta
ble
5C
orr
ela
tio
ns
of
Tra
dit
ion
-Ori
en
ted
Re
lig
iou
sne
ssa
nd
Sub
jec
tiv
eSp
irit
ua
lity
Wit
hSi
ng
leIt
em
s
Tradit.Religiousness
Subj.Spirituality
Measure
TR
SS
Religious
ESI-R
SDI
Alpha
Mystical
ESI-S
SDI
Delta
Hypothesized
tobeTR-related
Power
ofGod
.84n
.24
.74n
.79n
.82n
.17
.26n
.17
RespectReligiousLeaders
.67n
�.03
.60n
.59n
.67n
�.07
�.01
.00
RespectThose
WhoStudyScriptures
.67n
.08
.58n
.59n
.68n
.04
.10
.05
Power
ofMiracles
.62n
.33n
.50n
.61n
.61n
.21n
.37n
.21n
Power
ofSupernatural
.59n
.26n
.48n
.59n
.56n
.15
.32n
.16
RespectTen
Commandments
Obeyers
.53n
.00
.47n
.46n
.54n
.01
.05
�.05
RespectMilitary
Leaders
.35n
�.08
.31n
.24n
.43n
.00
�.10
�.11
RespectEvolutionScientists(�)
�.57n
.05
�.43n
�.47n
�.69n
�.02
�.03
.18n
RespectGayandLesbianPeople
(�)
�.37n
.19n
�.29n
�.25n
�.48n
.05
.15
.26n
RespectFem
inists
(�)
�.37n
.18n
�.29n
�.27n
�.46n
.08
.10
.26n
Hypothesized
tobeSS-related
Power
ofAstrologicalInfluences
.07
.55n
.03
.10
.06
.41n
.48n
.44n
Power
ofMagic
.03
.46n
�.02
.07
.04
.35n
.42n
.35n
RespectPsychics
�.04
.46n
�.05
.00
�.06
.27n
.39n
.43n
RespectEnlightened
Persons
.36n
.45n
.31n
.40n
.29n
.26n
.38n
.43n
Power
ofWitchcraft
.13
.31n
.10
.13
.12
.29n
.34n
.12
RespectUnconventionalBeliefs
�.05
.29n
�.05
.03
�.11
.12
.28n
.29n
Power
ofDestiny
.11
.30n
.05
.09
.18
.20n
.23n
.30n
Power
ofFate
�.01
.21n
�.07
�.05
.08
.13
.10
.28n
RespectAvoidersofMilitary
�.34n
.12
�.30n
�.23n
�.41n
.01
.16
.11
Hypothesized
toberelatedto
both
TR
andSS
Frequency
ofDivorces
�.03
.22n
�.02
�.02
�.06
.13
.19n
.21n
Number
ofBiologicalChildren
.13
.01
.16
.12
.09
.05
�.03
.01
Note.N
5375.TR—
Tradition-orientedReligiousnesscomposite,SS—
SubjectiveSpirituality
composite,ESI—
ExpressionsofSpirituality
Inventory,ESI-R—Religiousnessfactor,ESI-S—SpiritualExperiencesfactor,(�)indicatesthatanegativecorrelationwashypothesized.
npo.001.
to be SS-related—the power of witchcraft and fate, respect for those
with unconventional beliefs—had moderate correlations in theexpected direction but not consistently across all indicators. Respect
for avoiders of military service was more consistently related (butnegatively) to TR than to SS indicators. Overall, correlations in
Tables 3 through 5 suggest that SS involves an interest in privatelyand subjectively experienced (rather than collectively shared and
validated) metaphysical phenomena but that this interest does nothave substantial corollaries in the domain of political views.
TR and SS composites (put in standardized form) differed
among various denomination categories, which also were indexedby a single item. Among those categories with substantial (over
N5 25) representation in our sample, mean standard scores tiltedin the TR direction for Protestants (.49 for TR, � .15 for SS) and
Catholics (TR .50, SS .07). As would be expected, scores tilted inthe SS direction for those who identified as ‘‘spiritual but not
affiliated with a conventional religion’’ (TR � .31, SS .75). Thosewho endorsed ‘‘none’’ as their religious affiliation had below-mean
scores for both but were more extreme for TR (� 1.36) than for SS(� .60). Within a multivariate analysis of variance (overall F [6,634]5 58.27, po.001), denomination category (among these four,
with N5 321) predicted both TR (F [3, 317]5 105.33, po.001) andSS (F [3, 317]5 27.42, po.001). Post hoc Scheffe tests indicated
that, with 95% confidence intervals, Protestant and Catholic didnot differ from each other on either TR or SS but did differ from
both ‘‘spiritual’’ and ‘‘none,’’ just as ‘‘spiritual’’ and ‘‘none’’differed from each other, on both TR and SS. Overall, those who
identified with a religious denomination were much more likely tobe high on TR than were those who did not. SS, however, was auseful differentiator among the ‘‘unchurched’’; that is, it distin-
guished with very large effect size the ‘‘spiritual but not affiliatedwith a conventional religion’’ from those with ‘‘none’’ as religious
affiliation.
Correlations With Personality Dimensions
Table 6 presents correlations between the TR and SS indicators andbroad personality dimensions. The expected correlation of TR with
Agreeableness (A) and Conscientiousness (C) could not be foundwith the Big Five scales and only held up for Agreeableness when
1280 Saucier & Skrzypinska
Ta
ble
6C
orr
ela
tio
ns
of
Tra
dit
ion
-Ori
en
ted
Re
lig
iou
sne
ssa
nd
Sub
jec
tiv
eSp
irit
ua
lity
Wit
hB
igF
ive
an
dM
ult
i-L
an
gu
ag
eSe
ve
n
Tradit.Religiousness
Subj.Spirituality
Measure
TR
SS
TR0
SS0
Religious
ESI-R
SDI
Alpha
Mystical
ESI-S
SDIDelta
Big
FiveExtraversion
.02
.07
.02
.09
�.02
.06
.00
�.05
.08
.11
Big
FiveAgreeableness
.06
.13
.07
.17
.11
.12�.03
.03
.08
.20n
Big
FiveConscientiousness
.07�.15
.08�.19
.03
.07
.08
�.10
�.14
�.12
Big
FiveEmotionalStability
�.16�.15�.19�.20
�.11
�.17�.21n�.07
�.15
�.17
Big
FiveIm
agination/Intellect
�.11
.21n�.13�.28
�.09
�.02�.22n
.12
.28n
.07
NEO-PI-R
Extraversion
.05
.20n
.06
.25
.04
.11
.00
.07
.17
.23n
NEO-PI-R
Agreeableness
.29n
.13
.32
.16
.31n
.32n
.19n
.04
.07
.22n
NEO-PI-R
Conscientiousness
.07�.18
.08�.22
.07
.07
.05
�.11
�.17
�.15
NEO-PI-R
Neuroticism
.03
.07
.03
.09
.01
�.01
.10
.07
.07
.05
NEO-PI-R
Opennessto
Experience�.26n
.40n�.28
.49
�.19n
�.13�.37n
.24n
.33n
.37n
ML7Gregariousness
.06
.08
.07
.11
.04
.10
.03
�.05
.11
.12
ML7Self-Assurance
�.02
.00�.02
.00
�.03
.02�.05
�.03
�.01
.02
ML7Even
Tem
per
03�.07
.04�.09
.07
.05�.04
�.05
�.09
�.04
ML7Concern
forOthers
.28n
15
.35
.21
.25n
.30n
.24n
.08
.07
.21n
ML7Conscientiousness
.26n�.16
.32�.22
.23n
.20n
.28n�.11
�.12
�.14
ML7Originality/Intellect
�.13
.13�.16
.18
�.08
�.05�.22n
.11
.15
.03
ML7SocialUnacceptability(N
V)�.03
.28n�.04
.37
.05
�.03
.00
.23n
.30n
.16
Note.N
5375.TR—
Tradition-O
rientedReligiousnessfactor,SS—
SubjectiveSpirituality
factor,ESI—
ExpressionsofSpirituality
Inventory,
ESI-R—
Religiousnessfactor,ESI-S—
SpiritualExperiencesfactor,ML7—
Multi-LanguageSeven.
npo.001.
using the NEO-PI-R scales. However, TR was consistently moder-
ately correlated with ML7 Concern for Others (related to Agree-ableness) and ML7 Conscientiousness, in line with the hypothesis.
These results indicate that the correlation between TR and A may bestronger when the A measure emphasizes either compliance (as in the
NEO-PI-R) or prosocial and altruistic tendencies (as in the ML7),rather than gentleness and absence of hostility (as on the Big Five
scale). Similarly, the correlation between TR and C may be strongerwhen the C measure has greater emphasis on strictness and perfec-tionism, as is true of C in the ML7 (see Saucier, 2003, Table 4).
We expected that SS would be related to Openness to Experience(O) and to its lexical-factor counterpart Imagination/Intellect. We
found that O is related to both TR and SS and is a gooddifferentiator of them. Those high on TR tended to be low on O;
those high on SS tended to be high on O. This pattern generalizedweakly and inconsistently to the lexical factor in line with previous
findings that O is substantially related to social attitudes (McCrae,1996; van Hiel, Kossowska, & Mervielde, 2000), more so than is the
lexical Intellect factor (Yik & Tang, 1996).Finally, Table 6 reveals an interesting but unanticipated correla-
tion. Two of the SS indicators were correlated positively with Social
Unacceptability (i.e., Negative Valence), and the third was nearly so(po.01 but not po.001). There is some controversy over whether
Negative Valence (NV) has a substantive interpretation (Benet-Martinez & Waller, 2002; Saucier, 2002, 2003) or an artifactual
interpretation (Ashton & Lee, 2002). Saucier’s (2003) markers forthis factor include the adjectives weird and crazy; further analyses
indicated that individuals with strong tendencies toward subjective,mystical spiritual experiences were also more likely to apply theseparticular adjectives to themselves, consistent with a substantive
interpretation of NV and with the already noted correlation betweenSS and eccentricity.
Hierarchical Regression Analyses
Substantial correlations between personality scales and either TR or
SS raise the possibility that TR and SS are really ‘‘nothing morethan’’ traits already captured by personality inventories. The resolu-
tion of this issue depends, however, on which constructs one acceptsto be aspects of personality. A key question then is ‘‘How far must
1282 Saucier & Skrzypinska
one expand the definition of what comprises a personality variable in
order to argue that dispositions toward TR and SS are personalityvariables?’’
To answer this question, we employed a set of hierarchical-regression analyses. In each analysis the first block was the gender
variable. The second block of predictors consisted of lexical Big Fivescale scores. The third and fourth blocks consisted of other person-
ality scales that had demonstrated a substantial correlation with TRor SS indicators in the earlier analyses: the third block involved
scales whose categorization as ‘‘personality’’ is uncontroversial; thefourth block included scales whose categorization as such mightgenerate some debate. Finally, a fifth block involved scales that
would rarely be categorized as personality traits. The first fourblocks were utilized in a ‘‘forced entry’’ manner, whereas the fifth
block was subjected to a stepwise algorithm to search for the singlevariable that best added to the prediction (p for entry .001, p for
removal .005). As in earlier analyses, primary analyses involved theTR and SS composites, each based on three indicators, but we also
present analyses based on each indicator alone so as to examinegenerality of results across single indicators.
Table 7 summarizes the results of these hierarchical regressions.
Gender and the lexical Big Five each provided a significant R-squared change in only half of the regressions. Adding personality
inventory scales (for which there would be no dispute over theirbeing considered personality variables; e.g., Openness to Experience)
raised these multiple correlations substantially (to .41 and .61, with arange of .35 to .58 for the indicators). Adding still other scales that
might incite some controversy if labeled ‘‘personality’’ (e.g., Self-Transcendence) produced a very large increase in the multiple
correlation for TR (to .87, indicators in the .75–.85 range) and asmaller increase for SS (to .68, indicators in the .48–.69 range).
Switching to adjusted (shrunken) R values in these regressions
would result in a reduction of from .00 to .03 in the R values, exceptthat the R for Mystical and the lexical Big Five was adjusted from
.19 to .14. There is probably relatively little inflation in the multiplecorrelations.
The results of these four regression steps indicate that TR andSS should not be considered personality if personality means
‘‘captured by the lexical Big Five.’’ However, if personality is definedas ‘‘whatever is measured on influential personality inventories’’
Spiritual But Not Religious 1283
Ta
ble
7M
ult
iple
Co
rre
lati
on
sF
rom
Hie
rarc
hic
al
Re
gre
ssio
ns
Usi
ng
Pe
rso
na
lity
Sca
les
toP
red
ict
Re
lig
iou
s/Sp
irit
ua
lB
eli
ef
Dim
en
sio
ns
Tradition-O
rientedReligiousness
SubjectiveSpirituality
After
Step
TR
SS
Religious
ESIRelig.
SDIAlpha
Mystical
ESISpir.
SDIDelta
1—Gender
.17
.23n
.12
.21n
.14
.09
.21n
.26n
2—LexicalBig
Five
.27
.37n
.22
.27
.34n
.19
.40n
.37n
3—Other
Personality
Scales
.41n
.61n
.35n
.37n
.47n
.43n
.58n
.55n
4—Other
ScalesThatMightBeLabeled
‘‘Personality’’
.87n
.68n
.75n
.84n
.85n
.48n
.69n
.59n
5—Non-Personality
ScalesThatAddto
Prediction
.88n
.78n
.76n
.85n
.88n
.55n
.76n
.68n
Scale
Added
atStep5
RWA
IrrB
elRWA
Individualism
RWA
IrrB
elIrrB
elIrrB
el
ItsDirectionofEffect
11
1�
11
11
Note.N
5375.
nChanges
inR-squared
forstep
significant,
po.001.RWA—
RightWing
Authoritarianism.Individualism
—Oyserm
an
scale.IrrB
el—
IrrationalBeliefs.ForTRregressionsStep3included
NEO-PI-ROpennessto
Experience
andCPIVector2(N
orm
-favoring),Step4TCISelf-
Transcendence
andMPQ
Traditionalism
.ForSSregressionsStep3included
NEO-PI-R
Opennessto
Experience,MPQ
Absorption,60ML7
NegativeValence,Private
Self-Consciousness,Step4TCISelf-Transcendence,CESDissociation,CEQ
Fantasy-Proneness,Eccentricity.
(including the TCI and MPQ), then it appears that TR and SS
should definitely be considered aspects of personality.For all three TR indicators RWA had partial correlations in the
.15 to .45 range with the criterion after Block 4, but, in one case,Individualism had a higher (negative) one. For the SS indicators,
Irrational Beliefs was, for the aggregate as well as the singleindicators, the scale that added to prediction. Though Block 5
additions all yielded a significant change in R-squared values,many may be surprised that such belief scales added so little to
what can be called personality measures in terms of predictingvariation in tradition-oriented religiousness and subjective spirituality.
The foregoing results involved operationalizing TR and SS via
composites that allowed them to be modestly (.21) intercorrelated.However, these results were not method dependent. When analyses
were run using component or factor scores instead of composites,with either principal components or maximum-likelihood extraction
and with either oblique or with orthogonal factors, the coefficientswere little different and would lead to the same conclusions.
DISCUSSION
The results of this study indicated that Tradition-oriented Religious-
ness (TR) and Subjective Spirituality (SS) are highly independentdimensions that can be tapped by any of at least three sets of
indicators. Any of these indicators would have generated similarresults for most of our hypotheses, so these findings are not
contingent on whether the indicators are adjectives or, instead,factors drawn from the ESI or from studies of dictionary-based
isms. Caution is needed, however, in using the term spiritual insurvey items, as this term tends to lead to a confounding of TR and
SS, whereas terms like religious or mystical do not.These dimensions have quite different correlates. TR is highly
associated with authoritarianism and traditionalism and, more
moderately, with collectivism versus individualism and with (low)openness to experience; TR represents a brand of religious/spiritual
belief in which there is high reliance on tradition-hallowed sourcesof authority that provide shared practices (e.g., rituals) and rules
for controlling social and sexual behavior. SS is associated withabsorption, fantasy-proneness, dissociation, and beliefs of a magical
Spiritual But Not Religious 1285
or superstitious sort, as well as with eccentricity and high openness
to experience and represents a brand of belief in which the indivi-dual’s subjective experiences (including intuitions and fantasies)
have great importance.The two dimensions appear to be dispositions that are substan-
tially intertwined with commonly measured personality traits andshould not be confused with denominational affiliations, nor even
with membership in a particular religion. Perhaps such affiliationsand memberships reflect specific cultural and environmental influ-ences to which the individual might be exposed and thus involve only
the superficial mode in which these dispositions are expressed.However, those who identify with a denomination appear more
likely to be high on TR than are those who do not.Overall, our findings are potentially important for several reasons.
First, they indicate that scientists who treat religious/spiritualtendencies as a unitary phenomenon do so in error. Instead, spiritual
tendencies can go in either of two highly independent directions:toward a tradition-oriented, authority-based religion emphasizing
collectively shared beliefs, or toward a mysticism based in subjective,individual experience that seems to have little implication for groupaction or political views. Second, it appears that these two different
‘‘directions’’ are associated with different dispositions. One directionreferences collectively defined authority and provides clearer direc-
tives for behavior, making it more appealing to those whosetendencies and values emphasize behavioral control (via rules,
rituals, or a hierarchical conception of the world). The otherdirection, more subjective and phenomenological, may be more
appealing to those with tendencies to absorption and fantasy.Research is needed to define further these dispositions.
Arguments that religiousness/spirituality forms a ‘‘sixth factor’’ of
personality (e.g., MacDonald, 2000; Piedmont, 1999) should takeinto account that there are two relatively orthogonal dimensions in
this domain and that these two dimensions have substantial correla-tions (in opposing directions) with one of the widely accepted first
five factors, Openness to Experience. If one were to produce a singlereligiousness/spirituality factor that averaged TR and SS, it might
appear orthogonal to Openness. But as our findings demonstrate,TR and SS should not be lumped together.
One could, alternatively, attempt to assimilate these findingsentirely to the Five-Factor Model. This would involve the argument
1286 Saucier & Skrzypinska
that SS and TR simply represent high and low Openness to
Experience, respectively. A first problem with this approach: TRand SS are not opposites, but highly independent. This means some
individuals will be high on one and low on the other. But others willbe high on both and others will be low on both, and these two groups
of individuals might tend to score similarly—around the middle—onOpenness to Experience. Within the Five-Factor Model, information
that would distinguish high TR/high SS from low TR/low SS issimply lost. A second problem with this ‘‘Openness5SS–TR’’
formulation is that Openness has some problems with generalityacross cultural settings (De Raad, 1994; Watkins & Gerong, 1997).
We found that TR and SS are relatively independent of lexically
based personality factors and that lexically based social-attitude(isms) factors serve as good indicators for them. The clearest
approach might be to dispense with the Openness construct as aproblematic amalgam of temperamental, intellectual, and attitudinal
tendencies. Instead, one would conceive of TR and SS as disposi-tional factors underlying social attitudes that are beyond a Big Five
in which there is an Intellect factor rather than Openness.TR and SS dimensions do not apply only to religious people but
can also be used to differentiate among the nonreligious. For
example, one nonreligious person may strongly oppose traditionalreligion but be more indifferent to subjective spiritual experiences.
Another may be relatively indifferent to traditional religion butstrongly skeptical of subjective spiritual experiences.
Some limitations of this study need acknowledgment. First, we donot yet know to what extent our findings are culture bound. The
present sample represents but one cultural setting. Given thepresence in all major religions of distinct orthodox and mystical
schools, cross-cultural generalizability seems promising. However,relations with other variables may be moderated by culture. Forexample, if adherence to traditional authority-based religion were
strongly normative in a culture, endorsement of attitudes represent-ing doctrines of this religion should become highly desirable for
individuals. Under such conditions, measurement of tradition-orientedreligiousness should become strongly affected by desirability respond-
ing, as are scores on Agreeableness and Conscientiousness, with theresult that correlations of TR with these two variables should increase.
We relied on multiple converging indicators of TR and SS as thereis not yet a well-validated measure of these two constructs; ESI
Spiritual But Not Religious 1287
items, markers of Alpha and Delta isms factors, or adjectives all
seem capable of serving. Moreover, we relied exclusively on self-report data, though it can be argued that self-report may be superior
with respect to belief variables because informants often are not veryknowledgeable about the full range of another individual’s beliefs.
CONCLUSIONS
Individual differences in religious/spiritual beliefs cannot be cap-tured by a single dimension. Two highly independent dimensions
(TR and SS) have quite different correlates, supporting the view thatthey are indeed divergent constructs. TR is associated with author-
itarianism and traditionalism and, more moderately, with collecti-vism versus individualism and with low Openness to Experience. SSis associated with absorption, fantasy-proneness, dissociation, and
beliefs of a magical or superstitious sort, as well as eccentricity andhigh Openness to Experience. Expressions of religious/spiritual
belief appear to differ according to whether the person placesrelatively more importance on having clear collective standards for
behavioral control, or on personal intuitions, fantasies, and sub-jective experience. Because the two dimensions of religious/spiritual
belief overlap so substantially with personality, there is a case forsaying they are themselves aspects of personality. Even if they are
not personality, they appear to capture important dispositions of theindividual, dispositions to which psychology has paid too littleattention.
REFERENCES
Allport, G. W. (1959). Religion and prejudice. Crane Review, 2, 1–10.
Almagor, M., Tellegen, A., & Waller, N. (1995). The Big Seven model: A cross-
cultural replication and further exploration of the basic dimensions of natural
language of trait descriptions. Journal of Personality and Social Psychology, 69,
300–307.
Altemeyer, B. (1996). The authoritarian specter. Cambridge, MA: Harvard
University Press.
Argyle, M., & Beit-Hallahmi, B. (1975). The social psychology of religion. London:
Routledge.
Ashton, M. C., & Lee, K. (2002). Six independent factors of personality variation:
A response to Saucier. European Journal of Personality, 16, 63–75.
1288 Saucier & Skrzypinska
Batson, C. D., & Schoenrade, P. A. (1991). Measuring religion as quest: Validity
concerns. Journal for the Scientific Study of Religion, 30, 416–429.
Baumeister, R. F. (1991). Meanings of life. New York: Guilford Press.
Benet-Martınez, V., & Waller, N. G. (2002). From adorable to worthless: Implicit
and self-report structure of highly evaluative personality descriptors. European
Journal of Personality, 16, 1–41.
Bernstein, E. M., & Putnam, F. W. (1986). Development, reliability, and validity
of a dissociation scale. Journal of Nervous and Mental Disease, 174, 727–735.
Burris, C. T. (1994). Curvilinearity and religious types: A second look at intrinsic,
extrinsic, and quest relations. International Journal for the Psychology of
Religion, 4, 245–260.
Buss, A. H. (1980). Self-consciousness and social anxiety. San Francisco:
Freeman.
Church, A. T., Reyes, J. A. S., Katigbak, M. S., & Grimm, S. D. (1997). Filipino
personality structure and the Big Five model: A lexical approach. Journal of
Personality, 65, 477–528.
Cloninger, C. R., Przybeck, T., Svrakic, D., & Wetzel, R. (1994). The Tempera-
ment and Character Inventory (TCI): A guide to its development and use. St.
Louis, MO: Center for Psychobiology of Personality, Washington University.
Costa, P. T., & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO
PI-R) and NEO Five-Factor Inventory (NEO-FFI) professional manual.
Odessa, FL: Psychological Assessment Resources.
Crowne, D. P., & Marlowe, D. (1964). The approval motive: Studies in evaluative
dependence. New York: Wiley.
De Raad, B. (1994). An expedition in search of a fifth lexical factor: Key issues in
the lexical approach. European Journal of Personality, 8, 229–250.
Dollahite, D. C. (1998). Fathering, faith, and spirituality. Journal of Men’s
Studies, 7, 3–15.
D’Onofrio, B. M., Eaves, L. J., Murrelle, L., Maes, H. H., & Spilka, B. (1999).
Understanding biological and social influences on religious affiliation, atti-
tudes, and behaviors: A behavior genetic perspective. Journal of Personality,
67, 953–984.
Eckblad, M., & Chapman, L. J. (1983). Magical ideation as an indicator of
schizotypy. Journal of Consulting and Clinical Psychology, 51, 215–225.
Emmons, R. A. (1999). Religion in the psychology of personality: An introduc-
tion. Journal of Personality, 67, 873–888.
Fenigstein, A., Scheier, M. F., & Buss, A. H. (1975). Public and private self-
consciousness: Assessment and theory. Journal of Consulting and Clinical
Psychology, 43, 522–527.
Fuller, R. C. (2001). Spiritual, but not religious: Understanding unchurched
America. Oxford: Oxford University Press.
Gillings, V., & Joseph, S. (1996). Religiosity and social desirability: Impression
management and self-deceptive positivity. Personality and Individual Differ-
ences, 21, 1047–1050.
Goldberg, L. R. (1999). The Curious Experiences Survey, a revised version of the
Dissociative Experiences Scale: Factor structure, reliability, and relations to
demographic and personality variables. Psychological Assessment, 11, 134–145.
Spiritual But Not Religious 1289
Goldberg, L. R. (in press). The comparative validity of adult personality
inventories: Applications of a consumer-testing framework. In S. R. Briggs,
J. M. Cheek, & E. M. Donahue (Eds.), Handbook of adult personality
inventories. New York: Plenum.
Gough, H. G. (1996). CPI manual (3rd ed.). Palo Alto, CA: Consulting
Psychologists Press.
Greenberg, J., Pyszczynski, T., & Solomon, S. (1986). The causes and conse-
quences of a need for self-esteem: A terror management theory. In R. F.
Baumeister (Ed.), Public self and private self (pp. 189–212). New York:
Springer-Verlag.
Grice, J. W. (2001). Computing and evaluating factor scores. Psychological
Methods, 6, 430–450.
Kendler, K. S., Gardner, C. O., & Prescott, C. A. (1997). Religion, psychopathol-
ogy, and substance use and abuse: A multimeasure, genetic-epidemiological
study. American Journal of Psychiatry, 154, 322–329.
Koenig, H. G. (1997). Is religion good for your health? The effects of religion on
physical and mental health. New York: Haworth Pastoral Press.
Koopmans, P. C., Sanderman, R., Timmerman, I., & Emmelkamp, P. M. G.
(1994). The Irrational Beliefs Inventory: Development and psychometric
evaluation. European Journal of Psychological Assessment, 10, 15–27.
Kozielecki, J. (1991). Z Bogiem albo bez Boga [With God or without God].
Warszawa: PWN.
Krauss, S. (in press). The lexical structure of social attitudes in Romania: The isms
of Eastern Europe. Journal of Personality.
Kuczkowski, S. (1993). Psychologia religii [Psychology of religion]. Krakow:
WAM.
MacDonald, D. A. (2000). Spirituality: Description, measurement and relation to
the Five Factor Model of personality. Journal of Personality, 68, 153–197.
MacDonald, D. A. (2002). The Expressions of Spirituality Inventory: Test
development, validation and scoring information. Unpublished manuscript.
Ma)drzycki, T. (1996). Osobowosc jako system tworza)cy i realizujacy plany
[Personality as a system for creating and realizing plans]. Gdansk: GWP.
McCrae, R. R. (1996). Social consequences of experiential openness. Psychologi-
cal Bulletin, 120, 323–337.
McGuire, M. T., Troisi, A., Raleigh, M. J., &Masters, R. D. (1998). Ideology and
physiological regulation. In I. Eibl-Eibesfeldt & F. K. Salter (Eds.), Indoctrin-
ability, ideology, and warfare: Evolutionary perspectives (pp. 263–276). New
York: Berghahn.
Merkelbach, H., Horselenberg, R., & Muris, P. (2001). The Creative Experiences
Questionnaire: A brief self-report measure of fantasy-proneness. Personality
and Individual Differences, 31, 987–995.
Miller, W. R. (1998). Researching the spiritual dimensions of alcohol and other
drug problems. Addiction, 93, 979–990.
Newberg, A. B., & d’Aquili, E. B. (2000). The neuropsychology of religious and
spiritual experience. Journal of Consciousness Studies, 7, 251–266.
Newberg, A., Alavi, A., Baime, M., Pourdehnad, M., Santanna, J., & d’Aquili, E.
(2001). The measurement of regional cerebral blood flow during the complex
1290 Saucier & Skrzypinska
cognitive task of meditation: A preliminary SPECT study. Psychiatry Re-
search: Neuroimaging, 106, 113–122.
Norman, W. T. (1967). 2800 personality trait descriptors: Normative operating
characteristics for a university population. Department of Psychology,
University of Michigan.
Olson, J. M., Vernon, P. A., Harris, J. A., & Jang, K. L. (2001). The heritability of
attitudes: A study of twins. Journal of Personality and Social Psychology, 80,
845–860.
Oyserman, D. (1993). The lens of personhood: Viewing the self, and others, in a
multicultural society. Journal of Personality and Social Psychology, 65,
993–1009.
Paulhus, D. L. (1984). Two-component models of socially desirable responding.
Journal of Personality and Social Psychology, 46, 598–609.
Paulhus, D. L. (1988). Assessing self-deception and impression management in self-
reports: The Balanced Inventory of Desirable Responding. Unpublished manual,
University of British Columbia, Vancouver, Canada.
Piedmont, R. L. (1999). Does spirituality represent the sixth factor of personality?
Spiritual transcendence and the Five-Factor Model. Journal of Personality, 67,
983–1013.
Powell, L. H., Shahabi, L., & Thoresen, C. E. (2003). Religion and spirituality:
Linkages to physical health. American Psychologist, 58, 36–52.
Pratto, F., Sidanius, J., Stallworth, L. M., & Malle, B. F. (1994). Social
dominance orientation: A personality variable predicting social and political
attitudes. Journal of Personality and Social Psychology, 67, 741–763.
Sabatier, A. (1905). Religions of authority and the religion of the spirit
(L. S. Houghton, Trans.). New York: McClure, Phillips, & Co.
Saroglou, V. (2002). Religion and the five factors of personality: A meta-analytic
review. Personality and Individual Differences, 32, 15–25.
Saucier, G. (1994). Mini-Markers: A brief version of Goldberg’s unipolar Big Five
markers. Journal of Personality Assessment, 63, 506–516.
Saucier, G. (2000). Isms and the structure of social attitudes. Journal of
Personality and Social Psychology, 78, 366–385.
Saucier, G. (2002). Gone too far—or not far enough? Comments on the article by
Ashon and Lee (2001). European Journal of Personality, 16, 55–62.
Saucier, G. (2003). An alternative multi-language structure of personality attri-
butes. European Journal of Personality, 17, 179–205.
Saucier, G. (2004). Personality and ideology: One thing or two? Unpublished
manuscript, University of Oregon.
Saucier, G., & Goldberg, L. R. (1998). What is beyond the Big Five? Journal of
Personality, 66, 495–524.
Schmitt, N. (1996). Uses and abuses of coefficient alpha. Psychological Assess-
ment, 8, 350–353.
Schuon, F. (1953). The transcendent unity of religions (P. Townsend, Trans.). New
York: Pantheon.
Shafranske, E. P., & Gorsuch, R. L. (1984). Factors associated with the
perception of spirituality in psychotherapy. Journal of Transpersonal Psychol-
ogy, 16, 231–241.
Spiritual But Not Religious 1291
Skrzypinska, K. (2002). Poglad na swiat a poczucie sensu i zadowolenie z zycia
[View of the world, life meaning, and well-being]. Krakow: Impuls.
Skrzypinska, K. (in press). From spirituality to religiousness—Is this a one-way
direction? Anthology of Social and Behavioral Science. Linkoping, Sweden:
University of Linkoping.
Spilka, B., Hood, R. W., Hunsberger, B., & Gorsuch, R. (2003). The psychology of
religion: An empirical approach. New York: Guilford.
Tellegen, A. (in press).Manual for the Multidimensional Personality Questionnaire.
Minneapolis: University of Minnesota Press.
Thalbourne, M. A., & Delin, P. S. (1994). A common thread underlying belief in
the paranormal, creative personality, mystical experience and psychopathol-
ogy. Journal of Parapsychology, 58, 3–38.
Triandis, H. C., & Gelfand, M. J. (1998). Converging measurement of horizontal
and vertical individualism and collectivism. Journal of Personality and Social
Psychology, 74, 118–128.
van Hiel, A., Kossowska, M., & Mervielde, I. (2000). The relationship between
Openness to Experience and political ideology. Personality and Individual
Differences, 28, 741–751.
Vaughan, F. (1991). Spiritual issues in psychotherapy. Journal of Transpersonal
Psychology, 23, 105–119.
Waller, N. G., Kojetin, B. A., Bouchard, T. J., Lykken, D. T., & Tellegen, A.
(1990). Genetic and environmental influences on religious interests, attitudes
and values: A study of twins reared apart and together. Psychological Science,
1, 138–142.
Watkins, D., & Gerong, A. (1997). Culture and spontaneous self-concept among
Filipino college students. Journal of Social Psychology, 137, 480–488.
Wulff, D. M. (1997). Psychology of religion: Classic and contemporary (2nd ed.).
New York: Wiley.
Yik, M. S. M., & Tang, C. S. (1996). Linking personality and values: The
importance of a culturally relevant personality scale. Personality and Individual
Differences, 21, 767–774.
Zinnbauer, B. J., Pargament, K. I., Cole, B., Rye, M. S., Butter, E. M., &
Belavich, et al. (1997). Religion and spirituality: Unfuzzying the fuzzy. Journal
for the Scientific Study of Religion, 36, 549–564.
Zinnbauer, B. J., Pargament, K. I., & Scott, A. B. (1999). The emerging meanings
of religiousness and spirituality: Problems and prospects. Journal of Person-
ality, 67, 889–920.
1292 Saucier & Skrzypinska